Spinning parameter control method to improve the consistency of ultrafine fiber yarn

Abstract

The present application provides a spinning parameter control method for improving the consistency of ultrafine fiber yarns, and relates to the field of fiber yarn technology, including: determining multiple spinning processes based on target spinning equipment, and performing parameter identification on multiple spinning processes; introducing a spinning environment data set, and analyzing in combination with multiple spinning processes; combining multiple spinning parameters; performing reinforcement learning based on a spinning parameter group combined with a spinning environment data set, and dynamically updating multiple spinning parameters; performing simulation control on ultrafine fiber yarns, and formulating a control scheme to adaptively control multiple spinning parameters to improve the consistency of ultrafine fiber yarns. This application can solve the technical problem in the prior art that the lack of adaptive capability leads to the inability to respond to environmental changes in real time, resulting in poor consistency and stability of ultrafine fibers, achieve the technical goal of adaptive control, and achieve the technical effect of improving the stability and consistency of the ultrafine fiber production process.

Description

Technical Field

[0001] The present application relates to the field of fiber filament technology, and in particular to a spinning parameter control method for improving the consistency of ultrafine fiber filaments. Background Art

[0002] In the modern textile industry, the production technology of ultrafine fibers is gaining increasing attention because of its unique application potential in the fields of filtration, warmth preservation, medical materials, etc. However, current spinning technology still faces many challenges and defects in the production process of ultrafine fibers.

[0003] At present, existing spinning technologies often rely on fixed parameter settings and lack flexible adaptability. Although traditional wet spinning and dry spinning processes can control the diameter and strength of fibers to a certain extent, these parameters are usually determined by trial and error under laboratory conditions, and it is difficult to cope with complex and changeable environmental factors in actual production. For example, when the temperature or humidity changes during the production process, the originally set parameters may no longer be applicable, resulting in the production of substandard fibers. This lack of adaptive technology makes production efficiency low and it is difficult to ensure product consistency and stability.

[0004] In summary, the prior art has the technical problem of being unable to respond to environmental changes in real time due to lack of adaptive capability, which further affects the consistency and stability of the ultrafine fibers, not only reducing production efficiency but also leading to the production of unqualified products. Summary of the invention

[0005] The purpose of this application is to provide a spinning parameter control method for improving the consistency of ultrafine fiber yarns, so as to solve the technical problem that the prior art cannot respond to environmental changes in real time due to lack of adaptive capability, which further affects the consistency and stability of ultrafine fibers, not only reducing production efficiency but also leading to the production of unqualified products.

[0006] In view of the above problems, the present application provides a spinning parameter control method for improving the consistency of ultrafine fiber yarns, including: determining multiple spinning processes based on a target spinning equipment, performing parameter identification on the multiple spinning processes, and determining multiple spinning parameters; introducing a spinning environment data set, and analyzing the spinning environment data set in combination with the multiple spinning processes to determine a fiber evaluation index set; combining the multiple spinning parameters according to the fiber evaluation index set to determine a spinning parameter group; performing reinforcement learning based on the spinning parameter group in combination with the spinning environment data set, dynamically updating the multiple spinning parameters according to the learning feedback data, and determining the spinning optimization parameters; performing simulation control on the ultrafine fiber yarns according to the spinning optimization parameters, and formulating a control scheme to adaptively control the multiple spinning parameters, so as to improve the consistency of the ultrafine fiber yarns.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: By determining multiple spinning processes based on target spinning equipment, parameter identification is performed on the multiple spinning processes to determine multiple spinning parameters; a spinning environment data set is introduced, and a fiber evaluation index set is determined based on the spinning environment data set combined with the multiple spinning processes for analysis; the multiple spinning parameters are combined according to the fiber evaluation index set to determine a spinning parameter group; reinforcement learning is performed based on the spinning parameter group in combination with the spinning environment data set, and the multiple spinning parameters are dynamically updated according to the learning feedback data to determine the spinning optimization parameters; simulation control is performed on ultrafine fiber yarns according to the spinning optimization parameters, and a control scheme is formulated to perform adaptive control on the multiple spinning parameters to improve the consistency of ultrafine fiber yarns, and finally the technical goal of adaptive control is achieved, so as to achieve the technical effect of improving the stability and consistency of the ultrafine fiber production process.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 A schematic flow chart of a spinning parameter control method for improving the consistency of ultrafine fiber yarns in this application; Figure 2 A schematic diagram of the process of introducing a spinning environment data set into a spinning parameter control method for improving the consistency of ultrafine fiber yarns in this application. DETAILED DESCRIPTION

[0011] This application solves the technical problem that the prior art cannot respond to environmental changes in real time due to lack of adaptive capability, further affecting the consistency and stability of ultrafine fibers, not only reducing production efficiency, but also leading to the production of substandard products, by providing a spinning parameter control method for improving the consistency of ultrafine fiber yarns. The technical goal of adaptive control is achieved, and the technical effect of improving the stability and consistency of the ultrafine fiber production process is achieved.

[0012] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0013] Please see attached Figure 1 The present application provides a spinning parameter control method for improving the consistency of ultrafine fiber yarns, which specifically includes the following steps: Step 1: determining a plurality of spinning processes based on a target spinning device, performing parameter identification on the plurality of spinning processes, and determining a plurality of spinning parameters.

[0014] Specifically, the target spinning equipment is a machine for producing textiles, such as a rotary spinning machine. Through the analysis of the target spinning equipment, multiple spinning processes can be determined, for example, the spinning process includes wet spinning and dry spinning. Then, the key parameters required for each spinning process are identified, such as the polymer concentration in wet spinning is 20%, and then multiple spinning parameters, that is, the specific operating conditions of each spinning process, such as in wet spinning, the stretch ratio is 5:1, which is used to ensure that the quality and performance of the final textile meet the expected standards.

[0015] Step 2: introducing a spinning environment data set, analyzing the spinning environment data set in combination with the plurality of spinning processes, and determining a fiber evaluation index set.

[0016] Specifically, environmental parameters related to the spinning process, such as temperature, humidity, and air pressure, are collected. For example, the temperature is 25 degrees Celsius and the humidity is 50%, and a spinning environment data set is obtained. Subsequently, different process conditions are compared with the spinning environment data to identify the best production method, such as comparing the effects of wet spinning and dry spinning under the same environment, and then determining the fiber evaluation index set, including specific parameters such as fiber strength, toughness, and diameter, such as fiber strength reaching 12 MPa and diameter controlled within 300 nanometers, to ensure the production of high-quality textiles.

[0017] Step three: combining the multiple spinning parameters according to the fiber evaluation index set to determine a spinning parameter group.

[0018] Specifically, according to the pre-set quality standards and performance indicators, different spinning parameters are effectively matched and adjusted in order to achieve the best textile effect. For example, suppose the fiber evaluation index set includes parameters such as fiber strength, diameter consistency and toughness. By analyzing the combination of different parameters, a new spinning parameter group is generated to ensure that during the production process, the generated fiber not only meets the strength requirements, but also has good diameter consistency and toughness, thereby optimizing the spinning production and improving the overall quality of the final product.

[0019] Step 4: Perform reinforcement learning based on the spinning parameter group in combination with the spinning environment data set, dynamically update the multiple spinning parameters according to the learning feedback data, and determine the spinning optimization parameters.

[0020] Specifically, the spinning parameters (such as rotation speed and stretch ratio) are combined with environmental data (such as temperature and humidity) and these parameters are continuously optimized through machine learning methods. Reinforcement learning evaluates the effect of each parameter combination through a feedback mechanism. By analyzing the learning feedback data, it is possible to identify which parameter combinations perform best, thereby dynamically updating multiple spinning parameters. Ultimately, the optimal spinning parameters, i.e., the spinning optimization parameters, are determined to make the production process more efficient and the generated fibers meet the expected performance targets in terms of strength and toughness.

[0021] Step 5: Simulate and control the ultrafine fiber yarn according to the spinning optimization parameters, formulate a control scheme to adaptively control the multiple spinning parameters, and improve the consistency of the ultrafine fiber yarn.

[0022] Specifically, the fiber production process is simulated using the previously determined optimal parameter settings to evaluate its performance under different conditions. Then, based on real-time feedback and changes in the production environment, these parameters are dynamically adjusted to optimize the production process. For example, if temperature fluctuations are detected that cause fiber diameter instability, the control scheme may adjust the rotation speed and stretching ratio accordingly. Ultimately, adaptive control is used to improve the consistency of ultrafine fiber filaments, ensuring that each fiber maintains similar standards in terms of performance such as diameter and strength, thereby improving overall product quality.

[0023] The spinning parameter control method for improving the consistency of ultrafine fiber yarns can achieve the technical goal of adaptive control and achieve the technical effect of improving the stability and consistency of the ultrafine fiber production process.

[0024] Further, if Figure 2 As shown, this application also includes: Determine a plurality of sensor devices according to the operating parameters of the target spinning equipment, and construct a sensor network according to the plurality of sensor devices; perform real-time monitoring of the target spinning equipment through the sensor network to obtain a plurality of initial spinning environment data; smooth the plurality of initial spinning environment data according to the data acquisition sequence to generate a plurality of spinning environment data; cluster analyze the plurality of spinning environment data according to the plurality of spinning processes, classify the plurality of spinning environment data according to a plurality of clustering clusters, and obtain the spinning environment data set.

[0025] Specifically, multiple sensing devices are determined according to the operating parameters of the target spinning equipment, and are used to monitor different parameters through multiple sensors. For example, if the operating temperature of the equipment reaches 200 degrees Celsius, a temperature sensor is selected for real-time monitoring. Subsequently, multiple sensing devices are connected to form a unified sensor network to centrally collect data and ensure efficient information flow.

[0026] Through real-time monitoring of the target spinning equipment through the sensor network, multiple initial spinning environment data, such as temperature, humidity and pressure, can be obtained, providing a basis for subsequent analysis. For example, the monitored humidity is 60% and the temperature is 22 degrees Celsius. Among them, the humidity of the air affects the cooling rate and molding of the fiber. The air pressure has an impact on the extrusion and stretching of the fiber. The ventilation conditions in the workshop affect the cooling and molding of the fiber. The vibration of the spinning equipment and its surroundings affects the fiber formation process.

[0027] According to the data collection sequence, multiple initial spinning environment data are smoothed, and the collected data are filtered to eliminate noise and fluctuations, and multiple spinning environment data are generated to improve the accuracy of the data. For example, the humidity data over a period of time is corrected from 55%-65% to a stable 60% through smoothing, making the data more representative.

[0028] Cluster analysis is performed on multiple spinning environment data according to the characteristics of multiple spinning processes, that is, the environmental data is classified and processed to identify its influencing factors. For example, multiple spinning environment data are classified according to extrusion pressure, spinning speed, stretch ratio and cooling rate. For example, extrusion pressure affects the initial formation and diameter of the fiber; the faster the spinning speed, the faster the fiber cools, which in turn affects the diameter uniformity; the larger the stretch ratio during fiber stretching, the smaller the fiber diameter; the fiber cooling process directly affects its physical properties, including surface smoothness and consistency; the collection rate affects the final size and tensile stress of the fiber, thereby obtaining a spinning environment data set, which is helpful for subsequent process optimization.

[0029] By analyzing the operating parameters of the target spinning equipment, the appropriate sensing device is determined, and a sensor network is constructed. After real-time monitoring and acquisition of environmental data, a structured spinning environment data set is finally formed through smoothing and cluster analysis, which provides an important basis for optimizing the spinning process, thereby improving production efficiency and product quality.

[0030] Furthermore, the present application also includes: According to the multiple clusters, regression analysis is performed on the multiple spinning processes and the spinning environment data set to determine multiple correlation coefficients; based on the multiple correlation coefficients, multivariate prediction is performed on the multiple spinning parameters, and a parameter optimization set is determined according to the parameter prediction results, and the parameter optimization set includes a parameter optimization direction and a parameter optimization index; according to the parameter optimization direction and the parameter optimization index, consistency analysis of ultrafine fibers is performed to generate fiber analysis results; stability evaluation is performed on the fiber analysis results, and the fiber evaluation index set is constructed according to the evaluation results.

[0031] Specifically, multiple spinning processes and spinning environment data sets were regressed according to multiple clusters, and the relationship between different spinning processes and spinning environmental factors was explored through mathematical models to obtain the effectiveness of different processes in different environments. For example, when the humidity is 60%, the spinning intensity of the spinning process reaches 10 MPa, thereby determining multiple correlation coefficients to reflect the degree of correlation between the process and environmental factors.

[0032] Based on multiple correlation coefficients, multivariate prediction of multiple spinning parameters is performed, indicating that the results of regression analysis can be used to predict the performance of different parameters in actual production. For example, if the correlation coefficient between temperature and spinning strength is 0.85, it can be predicted that the strength value will increase by 10% at the set temperature. According to the parameter prediction results, the parameter optimization set is determined, including the parameter optimization direction (such as increasing or decreasing the temperature) and the parameter optimization index (such as the specific temperature value increased).

[0033] According to the parameter optimization direction and the parameter optimization index, the consistency analysis of the ultrafine fibers is carried out to analyze whether the generated ultrafine fibers maintain consistent characteristics. For example, if the fiber diameter is controlled at 500 nanometers through optimization, and the standard deviation of each sample is less than 50 nanometers, the consistency of the fiber is ensured, providing a basis for quality assessment.

[0034] By performing fatigue tests on samples, the stability of the fiber analysis results is evaluated, that is, whether the fiber's performance remains consistent over long periods of use, such as if the fiber's strength loss does not exceed 5% after 1,000 stretches. Based on the evaluation results, a set of fiber evaluation indicators is constructed to form a complete standard to measure the quality and performance of the fiber.

[0035] Through clustering and regression analysis, the relationship between spinning process and environmental data was obtained, and the optimal parameters were determined through multivariate prediction, thereby ensuring the quality of ultrafine fibers and providing a scientific basis for actual production.

[0036] Furthermore, the present application also includes: Traverse the multiple spinning parameters for initialization to determine an initial population; define a comprehensive fitness function based on the weighted sum of the fiber evaluation index set, calculate the initial population through the comprehensive fitness function to obtain a first fitness value, and the first fitness value corresponds to the initial population; perform random selection based on the initial population to determine N initial spinning parameters, scale and mutate the N initial spinning parameters to generate a first candidate solution set, where N is an integer greater than or equal to 3; cross-combine the first candidate solution set with the N initial spinning parameters to generate a second candidate solution set; calculate the second candidate solution set based on the comprehensive fitness function to obtain a second fitness value; compare and determine the first fitness value with the second fitness value, update the initial population according to the determination result, and obtain a spinning optimization parameter group.

[0037] Specifically, parameter combinations of all spinning parameters, such as rotation speed, stretch ratio and polymer concentration, are accessed in sequence to determine the initial population, that is, to generate initial parameter settings, such as a stretch ratio of 4:1 and a polymer concentration of 20%.

[0038] Next, a comprehensive fitness function is defined based on the weighted sum of the fiber evaluation index set to quantify the performance of each group of parameters. For example, the comprehensive fitness function weights the strength, toughness, and diameter indicators to obtain the first fitness value, which has a direct correspondence with the parameters in the initial population.

[0039] Then, random selection is performed based on the initial population to determine N initial spinning parameters, such as N is 5, and the extracted parameters may be a rotation speed of 3200 rpm, a stretch ratio of 3:1, etc. After scaling and mutating the N initial spinning parameters, that is, performing random adjustments, the first candidate solution set is generated, such as changing the stretch ratio to 4:1.

[0040] Next, the first candidate solution set is cross-combined with the N initial spinning parameters, i.e., reorganized, to obtain the second candidate solution set, such as a new rotation speed of 3100 rpm and a stretch ratio of 3.5: 1. The new individuals not only contain the information after mutation, but also retain some characteristics of the original individuals.

[0041] Subsequently, the second candidate solution set is calculated based on the comprehensive fitness function, that is, the new combination parameters are evaluated again to obtain a second fitness value, which may be 0.85.

[0042] Finally, the first fitness value is compared with the second fitness value, and the initial population is updated according to the comparison result. If the second fitness value is higher, the old parameters are replaced by the new parameters to obtain the spinning optimization parameter group, ensuring that the final parameter setting can maximize the fiber quality.

[0043] By traversing and optimizing multiple spinning parameters and combining fitness evaluation, the performance of parameter combinations is gradually improved to achieve the best spinning effect and fiber quality.

[0044] Furthermore, the present application also includes: Determine whether the second fitness value is greater than the first fitness value; if the second fitness value is greater than the first fitness value, generate an update instruction, use the update instruction to take the second candidate solution set as an index, traverse the initial population for update and replacement, and obtain the spinning optimization parameter group; if the second fitness value is less than or equal to the first fitness value, generate an iteration instruction, use the iteration instruction to randomly select the initial population and determine M initial spinning parameters; scale and mutate the M initial spinning parameters to generate a third candidate solution set, where M is an integer greater than or equal to 3, and the M initial spinning parameters are different from the N initial spinning parameters; cross-combine the third candidate solution set with the M initial spinning parameters to generate a fourth candidate solution set, and iteratively calculate the fourth candidate solution set through the comprehensive fitness function, compare the calculation result with the first fitness value and output a third fitness value; traverse the initial population according to the third fitness value for update and replacement to obtain the spinning optimization parameter group.

[0045] Specifically, by comparing the second fitness value with the first fitness value, it is determined whether the second fitness value is greater than the first fitness value to determine which parameter setting performs better. For example, if the first fitness value is 0.75 and the second fitness value is 0.85, the second fitness value is higher. If the second fitness value is greater than the first fitness value, an update instruction is generated to replace the old parameters, and the second candidate solution set is used as a new parameter set, thereby updating the initial population and obtaining a spinning optimization parameter group, such as adjusting the rotation speed to 3400 rpm.

[0046] If the second fitness value is less than or equal to the first fitness value, an iteration instruction is generated, indicating that the current parameter settings need to be further optimized. By randomly selecting the initial population, M initial spinning parameters are determined. Assuming that M is 4, the parameters that may be extracted are a stretch ratio of 3:1, etc. After scaling and mutating the initial spinning parameters, a third candidate solution set is generated, that is, the M initial parameters are adjusted, and the stretch ratio may be adjusted to 4:1.

[0047] The third candidate solution set is cross-combined with the M initial spinning parameters to generate a new fourth candidate solution set. For example, a new parameter setting is obtained through the combination, such as a rotation speed of 3300 rpm. Next, the fourth candidate solution set is iteratively calculated through the comprehensive fitness function. The iteration termination condition is to stop when the fitness is greater than the first fitness value. The calculation result is compared with the first fitness value to determine the output of the third fitness value. A new value of 0.78 may be obtained to evaluate the effect of the new parameters.

[0048] If the third fitness value is better, the new parameters are used to replace the old parameters, and the spinning process is continuously optimized to form an optimization cycle. The best individual in the current population is the optimal spinning parameter group, ensuring that the best spinning parameters are finally selected to improve production efficiency and product quality.

[0049] Furthermore, the present application also includes: Based on the spinning sequence, the spinning environment data set and the spinning parameter group are associated and combined to construct a spinning state space; based on the spinning sequence, multiple time steps are set, and the spinning parameter group is adjusted according to the multiple time steps to construct a spinning action space; the spinning state space is connected with the spinning action space, and the first spinning state parameter is extracted based on the spinning state space, and the first spinning action parameter is extracted based on the first spinning state and the spinning action space; based on the fiber analysis result, a reward function is constructed, and the first spinning state parameter and the first spinning action parameter are learned according to the reward function to obtain a state-action value, and the state-action value includes a state reward feedback value and an action reward feedback value; the state reward feedback value and the action reward feedback value are added to the learning feedback data.

[0050] Specifically, different environmental factors (e.g., temperature, humidity) and parameter settings (e.g., rotation speed, stretch ratio) were linked through time series analysis to construct the spinning state space as multidimensional data points.

[0051] Next, the entire spinning process is divided into several time periods. For example, the status is recorded every 10 seconds. The spinning parameter group is adjusted at each time step, and the operation of adjusting the process parameters (such as increasing the temperature, reducing the speed, etc.) is performed to form a spinning action space to respond to environmental changes.

[0052] The extracted state information is integrated with the corresponding action information, and the first spinning state parameter is extracted based on the spinning state space, such as the humidity is 60% at a certain moment, and then the first spinning action parameter is extracted based on the first spinning state combined with the spinning action space.

[0053] The performance evaluation during the spinning process (such as fiber strength, diameter consistency, etc.) is converted into a feedback mechanism as the basis for the learning algorithm. According to this reward function, the first spinning state parameter and the first spinning action parameter are learned to obtain a state-action value, including a state reward feedback value (such as a score based on strength assessment) and an action reward feedback value (such as a score based on the assessment effect of parameter adjustment).

[0054] Finally, all feedback results obtained during the learning process are summarized for subsequent optimization algorithm use, providing a data basis for the intelligent control system, thereby continuously improving spinning quality and efficiency.

[0055] By associating and combining the spinning environment data and parameter groups, constructing the state and action space, and further through the learning of the reward function, a complete feedback mechanism is formed to help optimize the spinning process and improve the fiber quality.

[0056] Furthermore, the present application also includes: Based on the state-action value, the multiple spinning parameters are traversed to determine multiple parameters to be optimized, and the multiple parameters to be optimized include spinning state parameters to be optimized and spinning action parameters to be optimized; based on the multiple time steps, the spinning state parameters to be optimized are dynamically updated according to the state reward feedback value to generate spinning state optimization parameters; based on the multiple time steps, the spinning state parameters to be optimized and the spinning action parameters to be optimized are dynamically updated according to the state reward feedback value to generate spinning state optimization parameters and spinning action optimization parameters; the spinning state optimization parameters are integrated with the spinning action optimization parameters to obtain the spinning optimization parameters.

[0057] Specifically, all parameter combinations are analyzed using feedback values ​​to identify parameters that need to be optimized. For example, the spinning state parameters to be optimized, such as humidity set to 55%, and the spinning action parameters to be optimized, such as the stretch ratio adjusted to 3:1. Then, the performance of the state parameters is evaluated at each time step and adjusted based on the feedback, such as updating the humidity from 55% to 60% to improve the spinning quality.

[0058] Then, the feedback of the action parameters is evaluated and adjusted. For example, the stretch ratio may be adjusted from 3:1 to 4:1 to generate the optimized parameters of the spinning state in order to obtain higher fiber strength. Finally, the adjusted humidity and stretch ratio are combined to form the final spinning optimization parameters to improve the overall production effect.

[0059] By continuously evaluating and adjusting the parameters to be optimized, we ensure that the spinning process can achieve the best results under different conditions, thereby improving the quality and performance of textiles.

[0060] Furthermore, the present application also includes: The ultrafine fiber filaments are simulated and controlled according to the spinning optimization parameters to generate a simulation control effect; the simulation control effect is detected based on the spinning environment data set to generate a plurality of simulation control detection results; the spinning optimization parameters are performance evaluated according to the plurality of simulation control detection results to generate a plurality of performance evaluation scores; an expected control threshold is set according to the plurality of performance evaluation scores; when the simulation control effect reaches the expected control threshold, the control plan is formulated based on the spinning optimization parameters.

[0061] Specifically, the fiber production process is simulated using optimized parameter settings such as temperature, rotation speed and stretch ratio to evaluate its performance under specific conditions. The corresponding fiber property data can be generated through simulation control.

[0062] Next, the simulation results were compared with the actual environmental conditions to verify their accuracy and reliability, and multiple simulation control test results were generated. For example, the fiber strength was measured to be 12 MPa under the same conditions, which was consistent with the simulation results. Then, the various test results were comprehensively considered to form a comprehensive evaluation system, thereby generating multiple performance evaluation scores, such as a strength score of 0.9 and a diameter consistency score of 0.85.

[0063] A target indicator is extracted from the evaluation results to guide subsequent production optimization. For example, the expected control threshold of fiber strength can be set to 0.85 to ensure that the produced fibers meet the standard in strength. Once the simulation results meet expectations, actual production processes and control measures are formulated based on the optimized parameters to ensure the best results in real production.

[0064] Through simulation control, detection and evaluation, the spinning parameters are continuously optimized to ensure that the quality and performance of the ultrafine fibers meet the predetermined standards, and ultimately achieve an efficient production control solution.

[0065] In summary, the spinning parameter control method for improving the consistency of ultrafine fiber yarns provided in the present application has the following technical effects: By determining multiple spinning processes based on target spinning equipment, parameter identification is performed on the multiple spinning processes to determine multiple spinning parameters; a spinning environment data set is introduced, and a fiber evaluation index set is determined based on the spinning environment data set combined with the multiple spinning processes for analysis; the multiple spinning parameters are combined according to the fiber evaluation index set to determine a spinning parameter group; reinforcement learning is performed based on the spinning parameter group in combination with the spinning environment data set, and the multiple spinning parameters are dynamically updated according to the learning feedback data to determine the spinning optimization parameters; simulation control is performed on ultrafine fiber yarns according to the spinning optimization parameters, and a control scheme is formulated to perform adaptive control on the multiple spinning parameters to improve the consistency of ultrafine fiber yarns, and finally the technical goal of adaptive control is achieved, so as to achieve the technical effect of improving the stability and consistency of the ultrafine fiber production process.

[0066] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0067] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A spinning parameter control method for improving the consistency of ultrafine fiber yarns, characterized in that: Methods include: Determine a plurality of spinning processes based on a target spinning device, perform parameter identification on the plurality of spinning processes, and determine a plurality of spinning parameters; Introducing a spinning environment data set, analyzing the spinning environment data set in combination with the plurality of spinning processes to determine a fiber evaluation index set; Combining the plurality of spinning parameters according to the fiber evaluation index set to determine a spinning parameter group; Perform reinforcement learning based on the spinning parameter group in combination with the spinning environment data set, dynamically update the multiple spinning parameters according to the learning feedback data, and determine the spinning optimization parameters; The ultrafine fiber filaments are simulated and controlled according to the spinning optimization parameters, and a control scheme is formulated to adaptively control the multiple spinning parameters, so as to improve the consistency of the ultrafine fiber filaments.

2. The spinning parameter control method for improving the consistency of ultrafine fiber according to claim 1, characterized in that: The spinning environment data set and the method include: Determining a plurality of sensor devices according to the operating parameters of the target spinning equipment, and constructing a sensor network according to the plurality of sensor devices; Real-time monitoring of the target spinning equipment is performed through the sensor network to obtain a plurality of initial spinning environment data; Smoothing the plurality of initial spinning environment data according to a data collection time sequence to generate a plurality of spinning environment data; The plurality of spinning environment data are clustered and analyzed according to the plurality of spinning processes, and the plurality of spinning environment data are classified according to a plurality of cluster clusters to obtain the spinning environment data set.

3. The spinning parameter control method for improving the consistency of ultrafine fiber according to claim 2, characterized in that: Analyzing the spinning environment data set in combination with the plurality of spinning processes to determine a fiber evaluation index set, the method comprising: performing regression analysis on the plurality of spinning processes and the spinning environment data set according to the plurality of clusters to determine a plurality of correlation coefficients; Performing multivariate prediction on the plurality of spinning parameters based on the plurality of correlation coefficients, and determining a parameter optimization set according to the parameter prediction results, wherein the parameter optimization set includes a parameter optimization direction and a parameter optimization index; Performing consistency analysis of ultrafine fibers according to the parameter optimization direction and the parameter optimization index to generate fiber analysis results; The fiber analysis results are subjected to stability evaluation, and the fiber evaluation index set is constructed according to the evaluation results.

4. The spinning parameter control method for improving the consistency of ultrafine fiber according to claim 1, characterized in that: Combining the plurality of spinning parameters according to the fiber evaluation index set to determine a spinning parameter group, the method comprising: Traversing the plurality of spinning parameters for initialization to determine an initial population; A comprehensive fitness function is defined based on the weighted sum of the fiber evaluation index set, and the initial population is calculated by the comprehensive fitness function to obtain a first fitness value, wherein the first fitness value corresponds to the initial population; Performing random selection based on the initial population to determine N initial spinning parameters, scaling and mutating the N initial spinning parameters to generate a first candidate solution set, where N is an integer greater than or equal to 3; Cross-combining the first candidate solution set with the N initial spinning parameters to generate a second candidate solution set; Calculating the second candidate solution set based on the comprehensive fitness function to obtain a second fitness value; The first fitness value is compared with the second fitness value, and the initial population is updated according to the determination result to obtain a spinning optimization parameter group.

5. The spinning parameter control method for improving the consistency of ultrafine fiber according to claim 4, characterized in that: The first fitness value is compared with the second fitness value, and the initial population is updated according to the determination result to obtain a spinning optimization parameter group, the method comprising: Determining whether the second fitness value is greater than the first fitness value; If the second fitness value is greater than the first fitness value, an update instruction is generated, and the second candidate solution set is used as an index through the update instruction to traverse the initial population for update and replacement to obtain the spinning optimization parameter group; If the second fitness value is less than or equal to the first fitness value, an iteration instruction is generated, and the initial population is randomly selected through the iteration instruction to determine M initial spinning parameters; Scaling and mutating the M initial spinning parameters to generate a third candidate solution set, where M is an integer greater than or equal to 3, and the M initial spinning parameters are different from the N initial spinning parameters; Cross-combining the third candidate solution set with the M initial spinning parameters to generate a fourth candidate solution set, iteratively calculating the fourth candidate solution set using the comprehensive fitness function, comparing the calculation result with the first fitness value, and outputting a third fitness value; The initial population is traversed and updated and replaced according to the third fitness value to obtain the spinning optimization parameter group.

6. The spinning parameter control method for improving the consistency of ultrafine fiber according to claim 3, characterized in that: Reinforcement learning is performed based on the spinning parameter group combined with the spinning environment data set, and the method includes: Associating and combining the spinning environment data set with the spinning parameter group based on the spinning time sequence to construct a spinning state space; Setting a plurality of time steps based on the spinning sequence, adjusting the spinning parameter group according to the plurality of time steps, and constructing a spinning action space; connecting the spinning state space with the spinning action space, extracting a first spinning state parameter based on the spinning state space, and extracting a first spinning action parameter based on the first spinning state combined with the spinning action space; Constructing a reward function based on the fiber analysis result, learning the first spinning state parameter and the first spinning action parameter according to the reward function to obtain a state-action value, wherein the state-action value includes a state reward feedback value and an action reward feedback value; The state reward feedback value and the action reward feedback value are added to the learning feedback data.

7. The spinning parameter control method for improving the consistency of ultrafine fiber according to claim 6, characterized in that: Dynamically updating the plurality of spinning parameters according to the learning feedback data to determine the spinning optimization parameters, the method comprising: Traversing the plurality of spinning parameters based on the state-action value to determine a plurality of parameters to be optimized, wherein the plurality of parameters to be optimized include a spinning state parameter to be optimized and a spinning action parameter to be optimized; Dynamically updating the spinning state parameters to be optimized according to the state reward feedback value based on the multiple time steps to generate spinning state optimization parameters; Based on the multiple time steps, the spinning state parameters to be optimized and the spinning action parameters to be optimized are dynamically updated according to the state reward feedback value to generate spinning state optimization parameters and spinning action optimization parameters; The spinning state optimization parameters are integrated with the spinning action optimization parameters to obtain the spinning optimization parameters.

8. The spinning parameter control method for improving the consistency of ultrafine fiber according to claim 1, characterized in that: The ultrafine fiber filaments are simulated and controlled according to the spinning optimization parameters, and the method includes: Performing simulation control on the ultrafine fiber filaments according to the spinning optimization parameters to generate simulation control effects; Detecting the simulation control effect based on the spinning environment data set to generate a plurality of simulation control detection results; Performing performance evaluation on the spinning optimization parameters according to the multiple simulation control detection results to generate multiple performance evaluation scores; setting a desired control threshold value according to the plurality of performance evaluation scores; When the simulation control effect reaches the expected control threshold, the control plan is formulated based on the spinning optimization parameters.

Images